Plasma-based untargeted metabolomics reveals potential biomarkers for screening and distinguishing of ovarian tumors
Ovarian cancer (OC), a leading cause of gynecological cancer mortality, is frequently detected at advanced stages due to asymptomatic early progression. This study investigates plasma-based untargeted metabolomics for identifying biomarkers to screen and differentiate ovarian tumors (OT). Plasma samples from OC, benign ovarian tumors (BOT), and healthy controls (HC) were analyzed. Samples were randomized into train and test sets, with differential metabolites screened via two-tailed Student's t-test and partial least squares discriminant analysis. ROC models evaluated discriminatory capacity. Key metabolites demonstrated high predictive value: TMAO and hippuric acid distinguished OT from HC (AUC > 0.95), while linoleic acid, alpha-linolenic acid, and arachidonic acid (AUC > 0.9) further supported OT screening. Kynurenine differentiated OC from BOT (AUC = 0.808). Reduced levels of specific lysophosphatidylcholines (LPC (17:0/0:0), LPC (15:0/0:0)) also distinguished OT from HC (AUC = 0.771-0.89). These findings suggest plasma metabolomics holds promise for noninvasive biomarker discovery in OT screening and distinguishing between malignant and benign cases, though further validation of metabolite quantification is warranted prior to clinical application.